[Endoscopic hemostasis in the treatment of bleeding peptic ulcer]

Upper gastrointestinal bleeding is still an important clinical problem, even though about 80% of bleedings due to peptic lesions stop spontaneously. Starting from 1970, a lot of endoscopic hemostatic procedures has been developed and employed with good results, but, at present, the key problem is represented by the selection and management of the patients at highest risk of rebleeding. This work reviews both techniques and devices most commonly used, but at present none of them can be considered actually better than the other ones. According to the SIED (Società Italiana di Endoscopia Digestiva) guidelines, the techniques of choice should be cheap, effective, safe and easy to perform, the pivotal role in determining the success rate of endoscopic hemostasis being played by the endoscopist's experience.

Sonographically guided biopsy and sonographic monitoring in the diagnosis and follow-up of 2 cases of sarcoidosis with hepatic nodules and inconclusive thoracic findings

OBJECTIVE

To report 2 cases in which abdominal sonography played a useful role in diagnosing sarcoidosis with early nodular hepatosplenic manifestations.

METHODS

In the first case, an asymptomatic woman with increased liver enzyme values underwent sonography, which showed multiple hypoechoic nodules in the liver and spleen. Computed tomography confirmed the hepatosplenic findings and showed micronodular infiltrates of both lung fields, without hilar and mediastinal lymphadenopathy. In the second case, in a woman with a cough, dyspnea, and increased liver enzyme levels, thoracic computed tomography showed right pleural effusion causing partial atelectasis of the lower and middle lobes and mediastinal lymphadenopathy.

RESULTS

Sonography and computed tomography showed multiple nodules of the liver and spleen and retroperitoneal lymphadenopathy. In both cases, bronchoscopy, bronchial and bronchioloalveolar lavages, and transbronchial and mediastinal biopsies had negative results. Sonographically guided biopsy of the hepatic nodules was carried out and sarcoid granulomas were detected in the portal areas. After 3 months of steroid therapy, liver enzyme values nearly normalized, and sonography showed the disappearance of all abdominal lesions. The first patient underwent control computed tomography, which confirmed the regression of the disease; the second patient refused control computed tomography.

CONCLUSIONS

Sarcoidosis can occur with atypical onset, and in selected cases sonography may play a useful role in its diagnosis and follow-up.

[Assessment of anxiety levels in patients during diagnostic endoscopy]

Patients with increased pre-endoscopic anxiety take advantage of the use of conscious sedation. However, the habit of using premedication varies according to the type of endoscopic examination. Aims of this study were: to evaluate whether different endoscopic procedures may have different effects on patient anxiety; and whether anxiety affects patient tolerance. 163 consecutive outpatients undergoing endoscopy (75 gastroscopy, 51 colonoscopy and 37 bronchoscopy) were interviewed to evaluate pre-endoscopy anxiety, by using the Spielberger State-Trait Anxiety Inventory. After endoscopy, endoscopists rated patient cooperation on a 10-cm visual analogue scale. Females had state and trait anxiety levels higher than males (50.49 +/- 1.37 vs 45.07 +/- 1.20, p < 0.01; and 44.46 +/- 1.06 vs 38.48 +/- 1.01, p < 0.01). Anxiety levels were not related with the type of endoscopic procedure. A direct relationship was observed between state and trait anxiety (p < 0.001). Patient cooperation to endoscopy was inversely related with both trait (p < 0.05) and state anxiety (p < 0.01). In conclusion, gastroscopy, colonoscopy and bronchoscopy seem to induce similar pre-endoscopy anxiety levels. Therefore, the different frequency in the use of intravenous sedation between bronchial and gastrointestinal endoscopy does not seem justified.

[Endoscopic treatment of upper gastrointestinal non-variceal bleeding]

BACKGROUND

Endoscopic hemostasis is the method of choice for the treatment of bleeding peptic ulcers. This retrospective study was carried out to evaluate its effectiveness in routine endoscopic practice.

METHODS

The records of all patients with gastrointestinal bleeding undergoing esophagogastroduodenoscopy (EGDS) at the Endoscopy Center of Ferrara in the last year were retrospectively evaluated. For each patient with peptic lesions, the following data were recorded: demographic characteristics, use of NSAIDs, co-morbidity, hemodynamic conditions, blood transfusions before EGDS, time between onset of symptoms and EGDS, endoscopic findings, method of endoscopic haemostasis carried out, Rockall score and outcome. If re-bleeding occurred, the data concerning the second therapeutic intervention were recorded as well.

RESULTS

Seventy-six males and 45 females (age 34-92 years) entered the study. In 22% of cases no co-morbidity was present. Active bleeding was observed in 38% of cases, features consistent with recent bleeding in 54% of cases, and no sign of bleeding in 6.6% of cases. Peptic ulcer was observed in 89% of cases. Hemostasis was carried out in 81 patients (76 had adrenalin infiltration, 2 had argon plasma coagulator [APC], and 3 had both treatments); 26 patients had re-bleeding. The probability of re-bleeding was related to female gender (p<0.05; OR: 3.74), time between onset of symptoms and EGDS >24 hours (p<0.01; OR: 8.67), and presence of non-ulcer peptic pathology (p<0.05; OR: 0.15). Seven re-bleeding patients underwent surgery, 19 had endoscopic treatment. In 11 of these patients second hemostasis was resolutive, 8 bled again.

CONCLUSIONS

Endoscopic hemostasis of bleeding peptic lesions is effective also in routine clinical practice. Adrenalin infiltration is safe, easy to perform, cheap and repeatable, and in our opinion it should be considered the technique of first choice in endoscopic hemostasis.

Evaluation of a standardized protocol of intracavitary recombinant interferon alpha-2b in the palliative treatment of malignant peritoneal effusions. A prospective pilot study

OBJECTIVE

Several schedules with variable doses of intracavitary interferon have been proposed for the management of metastatic peritoneal effusions. This prospective pilot study evaluated the efficacy of a standardized schedule of intraperitoneal interferon alpha(2b).

METHODS

In 41 cancer patients with malignant ascites a 9-french intraperitoneal catheter was placed under sonographic guidance, and ascites was drained until abdominal ultrasound showed complete absence of effusion. Interferon alpha(2b), 6 or 9 million units (body weight <or= or >50 kg, respectively), was then administered via the tube, which was clamped for 6 h. 6 courses were given at 4-day intervals. In comparison with pretreatment levels, the 30-day response was classified as complete (CR) = no fluid recurrence, partial (PR) = fluid recurrence <50% and no response (NR) = >50%. Responders were monitored until fluid recurrence requiring paracentesis.

RESULTS

12 patients had CR, 15 PR, 14 NR. Global response (GR) was 65.9%. In ovarian cancer, GR was 75% and was not influenced by the quantity of the pretreatment daily fluid production. In the other cancers, a pretreatment fluid production <or=500 ml/day was associated with a significantly higher response (p < 0.05). Patients not showing response within the first 3 courses did not respond after the next courses. 60-day GR was 41.5%, 90-day GR was 26.8%. The mean time to fluid reaccumulation was 70.5 +/- 75.3 versus 11.4 +/- 4.4 days before treatment (p < 0.001). The survival of responders was 38-394 days (mean 131), and the average duration of response was 101 days.

CONCLUSION

This protocol could be reasonably applied to ovarian cancer patients and to patients with other cancers having a fluid production of <or=500 ml/day, and it should be stopped if no response is observed after 3 courses of treatment.

Distal trisomy of 10q. Report of a new case of duplication 10q25.2-25.3-->qter defined by FISH

In the present work, we report on a 2.5-year-old male patient with typical clinical features of partial trisomy of the distal third of chromosome 10 long arm. The karyotype was: 46,XY, dir dup(10)(q25.2-25.3-->qter). The identification of the duplicated segment was carried out by the fluorescence in situ hybridization technique using region-specific probes. The proband's phenotype is compared with previously reported cases.

Appropriateness of upper gastrointestinal endoscopy: a hospital-based study

Aims of this study were to evaluate: (1) whether upper gastrointestinal endoscopy (UGE) is used appropriately according to the American Society for Gastrointestinal Endoscopy (ASGE) and British Society of Gastroenterology (BSG) guidelines in a hospital setting and (2) whether there is any relationship between appropriateness of UGE and the presence of lesions detected by endoscopy. Indications and endoscopic findings for 734 consecutive UGE performed in 697 inpatients were retrospectively evaluated using ASGE and BSG guidelines to determine appropriateness of referrals. UGE showing endoscopic findings that had direct therapeutic or prognostic consequences were classified as "positive"; the other UGEs were classified as "negative." In all, 46% of UGEs were "positive," 54% "negative," and 61.7% and 23.2% of UGEs were inappropriate according to ASGE and BSG guidelines, respectively (P < 0.001). The probability of finding a positive endoscopy was significantly higher in UGE rated as appropriate than in those rated as inappropriate on the basis of ASGE guidelines (P < 0.001), but not on the basis of BSG guidelines. Endoscopies rated as inappropriate according to ASGE and BSG criteria showed a positive finding in 37.3% and 42.3% of cases, respectively (not significant difference). Multivariate analysis showed that the positive finding is directly related to age (P < 0.05), male gender (P < 0.001), prior UGE (P < 0.05), hematemesis (P < 0.001), and inversely related with upper abdominal pain (P < 0.01) and dyspepsia (P < 0.05). In hospitalized patients, UGE is frequently used for inappropriate indications, according to both ASGE and BSG guidelines. However, the actual clinical usefulness of appropriateness criteria, such as those proposed by ASGE and BSG, is questionable, as their strict observance could lead to missing a large number of significant endoscopic findings.

Increased myeloperoxidase index and large unstained cell values can predict the neutropenia phase of cancer patients treated with standard dose chemotherapy

OBJECTIVE

The objective of this study was to better understand neutropenia induced by standard dose chemotherapy and to verify if there are any hematological parameters for defining the phase and possibly the duration of neutropenia.

METHODS

The kinetics of large unstained cells (LUCs) and lymphocytes was evaluated in 324 blood counts of 56 chemotherapy cycles through the use of a Technicon H2 or an ADVIA 120 hematology analyzer. Blood samples collected during the neutropenia phase were also studied by flow cytometry using a large panel of monoclonal antibodies. Parametric and nonparametric statistics were employed to compare the different variables analyzed. A linear regression between each variable before and after nadir and a simple linear correlation among the same variables in the neutropenic and recovery phase were performed.

RESULTS

The percentage of LUCs reaches the higher value at nadir and the difference between the mean value of prenadir and nadir is statistically significant (P <.01). The number of LUCs increases during the pre and postnadir phase. Lymphocytes number appears stable in the prenadir phase. The MPXI index increases in the prenadir phase and falls at nadir and this difference is statistically significant(P <.01). LUCs are correlated with blasts and CD34+ cells in the pre and postnadir phase, with CD3+/CD4+ cells in the prenadir phase, and with CD2+/CD56+ in the postnadir phase.

CONCLUSIONS

Our data have shown that the estimation of both percentage of LUCs and MPXI can predict the neutropenia phase and orient for its duration. The lymphocyte number may be regarded as a parameter of risk of fever after day 5 of chemotherapy and the number of blood CD34+ cells may be predicted by LUC count.

[Ocular ultrasonography in pediatrics: persistence of hyperplastic primary vitreous]

AIM OF THE STUDY

To assess the diagnostic viability of ultrasound and color Doppler ultrasound in a particular segment of paediatric ophthalmology-persistent primary hyperplastic vitreous, in the presence of leucocoria, retrospectively evaluated in patients observed over the last two years.

MATERIAL AND METHODS

We re-evaluated four patients (two new-born, one unweaned and one nine-years-old) who at ophthalmoscopic examination were suspected having persistent primitive hyperplastic vitreous. The follow-up included an ultrasound examination, basic color-Doppler ultrasound and re-evaluation during sedation both using an operative microscope and ultrasound. The examinations were performed with linear 7.5 Mhz probes and a Doppler frequency of 3.7 Mhz, while the sedation examination was performed with an anular 13 Mhz probe. The scans were sagittal and axial to correctly localize the lesion with respect to the lens, to ciliary bodies and to the optic nerve head.

RESULTS

In the first patient ultrasound revealed a hyperechogenic inhomogeneous structure bilaterally in the vitreous structure; this extended from the posterior wall of the lens to the optic nerve head and retina, and was found to be highly vascolarized at the subsequent color-Doppler ultrasound. In the second patient there was an echogenic band extending from the posterior wall of the lens to the optic nerve head and to the retina, as the fourth patient showed a series of echogenic bands extending from the temporal ciliar of bodies to the temporal retina; in both cases no significant vascolarization was found by color Doppler ultrasound. In the third patient ultrasound showed a lesion involving nearly all of the vitreous body, with inhomogeneous structure with small calcifications and intense vascolarization.

DISCUSSION AND CONCLUSIONS

Ultrasound supplies essential information for the diagnosis of persistent hyperplastic primary vitreous as it determines the presence of the lesion, its extension and retinal and optic nerve head involvement. As confirmed by operative microscopy the first patient had bilateral lesions involving the optic nerve head, as well as the retina, the latter which appeared raised. The second and fourth patients had typically posterior lesions; the lesion of the third patient was difficult to interpret even by operative microscopy. To conclude,Ultrasound proved to have an optimal sensitivity and specificity for precisely locating the site and extension of the lesion. By contrast, the difficulties and need for sedation with color Doppler ultrasound unlikely to be useful with paediatric patients.

Randomized trial of omeprazole or ranitidine versus placebo in the prevention of chemotherapy-induced gastroduodenal injury

PURPOSE

Anticancer drugs may induce acute mucosal injury to stomach and duodenum. This study was planned to evaluate the efficacy of omeprazole or ranitidine in preventing such an injury.

PATIENTS AND METHODS

Two hundred twenty-eight cancer patients with normal stomach and duodenum or with less than three erosions, who were selected to be treated with cyclophosphamide, methotrexate, and fluorouracil (90 breast carcinoma patients) or fluorouracil alone (138 colon carcinoma patients), were randomly assigned to treatment with omeprazole 20 mg, ranitidine 300 mg, or one placebo tablet a day. Seven days after the second course of chemotherapy (CT), the patients underwent a further esophagogastroduodenoscopy to evaluate the mucosal injury. Endoscopic findings were quantified on the basis of an arbitrary score, and the occurrence of epigastric pain or heartburn was assessed weekly.

RESULTS

A significant difference was found among the three groups (P =.0032), as well as between pre- and postCT endoscopic findings (P =.00001). Endoscopic scores after CT were significantly higher than pretreatment scores in the placebo (P =.003) and ranitidine (P =.003) groups but not in the omeprazole group (P =.354). Acute ulcers were significantly less frequent in patients receiving omeprazole or ranitidine than in those receiving placebo (P =.0001 and P =.0315, respectively). Epigastric pain and/or heartburn were significantly less frequent in patients receiving omeprazole (P =.00124) or ranitidine (P =.038) than in those receiving placebo.

CONCLUSION

Omeprazole is effective in preventing chemotherapy-induced gastroduodenal injury. Ranitidine is effective in reducing the frequency of ulcers and upper gastrointestinal symptoms but is not effective in preventing the global endoscopic worsening caused by chemotherapy. The different efficacy of omeprazole and ranitidine can be explained by their different pharmacodynamics.

Subcutaneous seeding after ultrasound-guided placement of intrapleural catheter. An unusual complication of the intracavitary palliative treatment of pleural mesothelioma

Intrapleural catheters are useful in the palliative treatment of malignant effusions. Complications are infrequent and of little importance. We report a case of subcutaneous implantation metastasis along the course of intrapleural catheter, which had been placed under sonographic guidance in a patient with pleural mesothelioma. After drainage of the effusion, cisplatin plus cytarabine was administered via the chest tube, achieving complete remission of the pleural effusion. Subcutaneous metastasis became evident 3 months later and was the only sign of disease progression for 2 months. The seeding of cancer cells was probably caused by a small leakage of fluid around the chest tube that occurred during the placement procedure as a result of the increased intrapleural pressure caused by the large quantity of fluid that had accumulated in the pleural space.

Neuron-specific enolase, thymidine kinase, and tissue polypeptide-specific antigen in diagnosis and response to chemotherapy of small-cell lung cancer

The clinical usefulness of neuron-specific enolase (NSE), thymidine kinase (TK), and tissue polypeptide-specific antigen (TPS) was investigated in 41 patients (53-80 years old) with recently discovered small-cell lung cancer (SCLC). Eleven patients exhibited limited disease (LD) and 30 extensive disease (ED). Serum samples for NSE, TPS (immunoradiometric assay), and TK (radioenzymatic assay) evaluations were drawn from all patients at the time of diagnosis and before each cycle of chemotherapy in the treated patients. Therapy consisted of i.v. carboplatin 300 mg/m2 on the first day and i.v. etoposide 120 mg/m2 from the first to the third day every 3 weeks. Nine patients refused or were not eligible for chemotherapy. Five patients received only one course and showed no response (NR); 9 patients received two courses; 18 patients received three or more courses. In the last group, complete remission (CR) was obtained in 9 cases, partial remission (PR) in 18 cases. The tumor markers studied did not show any significant difference in distinguishing LD from ED. NSE and TPS were significantly more often abnormal than TK, either at the time of diagnosis (p < 0.05) or in PR or NR patients (p < 0.05). In relation to chemotherapy response, NSE and TPS serum patterns were shown to be more reliable than TK in PR (p < 0.05) and NR patients (computed error between 10% and 15%). No significant difference was observed between serum NSE and TPS patterns. Serum NSE and TPS seem to be more useful in the diagnosis and follow-up of SCLC patients undergoing chemotherapy. Further trials are necessary to ascertain whether the associated assessment of NSE and TPS can add useful information to that provided by the assessment of NSE alone.

Evaluation of a new ultrashort triple therapy for Helicobacter pylori disease

BACKGROUND

1-week proton pump inhibitor-based triple therapies are considered the most effective and convenient drug combinations for curing Helicobacter pylori infection. Short therapies, lasting less than 1 week have been investigated rarely.

AIM

To assess the efficacy and tolerability of a 3-day lansoprazole triple therapy after 1 day of lansoprazole pre-treatment.

METHODS

Seventy H. pylori-positive (rapid urease test and histology) patients received LAzT3 regimen (lanzoprazole 30 mg b.d. and azithromycin 500 mg o.m. for 3 days; tinidazole 2000 mg o.m. on day 1 and 1000 mg o.m. on days 2-3) after 1 day of lansoprazole pretreatment. Patients with active ulcer received lansoprazole 30 mg o.m. for an additional 4 weeks. Follow-up gastroscopy was carried out 4-6 weeks after completion of therapy. Eradication was defined as negative histology and rapid urease test.

RESULTS

Four patients failed to attend the follow-up endoscopy. One patient complained of minor side-effects. H. pylori was eradicated in 57 of 66 patients suitable for evaluation, with a per-protocol cure rate of 86.3% (95%CI: 76-94%), and an intention-to-treat cure rate of 81.4% (95%CI: 70-90%).

CONCLUSIONS

This new ultrashort triple therapy including lansoprazole, azithromycin and tinidazole seems to be effective in eradicating H. pylori. It is safe and well-tolerated, and may be taken into consideration as a valid alternative to the better known and widely used 1-week proton pump inhibitor-based triple therapies.

Usefulness of brushing urease test for diagnosis of Helicobacter pylori infection

BACKGROUND AND AIMS

Gastric brushing cytology is an accurate technique for rapidly detecting Helicobacter pylori infection, but it is not routinely employed since the presence of personnel experienced in this field, is necessary in the endoscopy suite. To evaluate the diagnostic usefulness of rapid urease test carried out on cytological brushing.

PATIENTS

A total of 143 consecutive patients with suspected Helicobacter pylori infection, referred for elective gastroscopy.

METHODS

For each patient, 2 brushings were carried out and 4 biopsies were taken from antral mucosa during gastroscopy. The former brushing was smeared on a slide, and stained by a rapid staining set for blood smears. The latter was used for rapid urease test, by shaking the brush into the urea broth. Two biopsies were used for rapid urease test and two for histologic examination. Histology was considered as the gold standard.

RESULTS

Of 143 patients, 73 were diagnosed as Helicobacter pylori infected using histology. Six brushing slides were inadequate due to insufficient cytology material. Biopsy-rapid urease test and brushing-rapid urease test had similar sensitivity (87.3% vs 83.5%), specificity (98.4% vs 96.8%) and overall accuracy (92.3% vs 89.5%). In 62 Helicobacter pylori infected patients, both rapid urease test techniques were positive. Brushing-rapid urease test became positive in a significantly shorter time than biopsy-rapid urease test (22 +/- 54 minutes vs 39 +/- 63 minutes; p < 0.01).

CONCLUSIONS

Brushing-rapid urease test is as accurate as biopsy-rapid urease test in detecting Helicobacter pylori infection, but it is significantly faster. Special care should be taken to carry out brushing adequately, to minimize the occurrence of false negatives.

A four-day low dose triple therapy regimen for the treatment of Helicobacter pylori infection

OBJECTIVE

The current guidelines recommend 1-wk triple therapy regimens for eradicating H. pylori infection. Until now, shorter regimens have scarcely been investigated. Azithromycin is a new generation macrolide antibiotic with unusual and favorable pharmacokinetics, and seems to be a very promising agent for innovative anti-H. pylori regimens. We assessed the efficacy and tolerability of a new 4-day low dose triple therapy in comparison with a well established 1-wk triple therapy in the treatment of Helicobacter pylori infection.

METHODS

One hundred-sixty consecutive patients with biopsy-proven H. pylori infection were randomized to receive lansoprazole 30 mg b.i.d. on days 1-4, azithromycin 500 mg u.i.d. on days 2-4, and tinidazole 2000 mg u.i.d. on day 3 (LAT group), or 7 days of triple therapy of omeprazole 20 mg u.i.d., clarithromycin 250 mg b.i.d., and tinidazole 500 mg b.i.d. (OCT group). Patients with gastric or duodenal active ulcer received proton pump inhibitors for an additional 4 wk. H. pylori eradication was defined as negative of both rapid urease test and histology on biopsies taken from the gastric body and antrum at least 1 month after the end of treatment.

RESULTS

Seven patients in the LAT group and four in the OCT group were lost to follow-up. No significant difference in either efficacy or tolerability was observed between the two regimens. Active ulcers healed in 97.8% of cases with LAT and in 100% of cases with OCT. The eradication rate was 80.8% in the LAT group and 85.5% in the OCT group, considering the per-protocol results, and 73.3% and 81.2%, respectively, considering the intention-to-treat results. Side effects occurred in one LAzT patient and in two OCT patients; they were mild and did not interfere with compliance.

CONCLUSION

The new proposed ultrashort triple therapy, including lansoprazole, low dose azithromycin for 3 days, and a single dose of tinidazole, appears to be a very effective anti-H. pylori regimen, a simpler, cheaper, well-tolerated, and equally effective alternative to 1-wk triple therapy.

Touch cytology. A reliable and cost-effective method for diagnosis of Helicobacter pylori infection

A variety of reliable methods are available for detecting Helicobacter pylori (Hp) during upper gastrointestinal endoscopy. We evaluated the clinical utility and cost-effectiveness of rapid urease test (RUT), touch cytology (TC), and histology (H). Two hundred thirty-eight consecutive patients (178 without previous medical treatment and 60 formerly treated with anti-Hp therapy) were tested for Hp infection by RUT, TC, and H (H&E stain). The infection status for each patient was established by a concordance of two test results. The time to carry out the three tests and their cost were also calculated. Sensitivity of TC (100%) was significantly higher than that of RUT (86.8%; P < 0.001), but not than that of H (94.9%). RUT was significantly more specific than H (100% vs 95.6%; P < 0.05), but not than TC (96.4%). Hp infection was more frequent in the patients with chronic active gastritis than in those with chronic nonactive gastritis (P < 0.001). No Hp infection was detected in absence of chronic antral inflammation. RUT resulted the cheapest method and H the most expensive; TC is faster and cheaper than H. When additional information about the severity of mucosal damage or the presence of cell atypias is not necessary, histologic examination can be omitted, and a cost-effective strategy for assessing Hp status might consist in taking two antral biopsies, the former for performing RUT and the latter for preparing a slide by TC, which should be stained and examined only when the RUT result is negative.

Misoprostol and omeprazole in the prevention of chemotherapy-induced acute gastroduodenal mucosal injury. A randomized, placebo-controlled pilot study

BACKGROUND

Chemotherapy (CT) may induce acute mucosal injury to the stomach and duodenum, but its prevention has been scarcely investigated.

METHODS

One hundred and eighty-two cancer patients with normal stomach and duodenum or having fewer than 3 erosions, selected to be treated with cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) (77 breast carcinoma patients) or 5-fluorouracil (5-FU) (105 colon carcinoma patients), were randomly assigned to prophylactic treatment with misoprostol, 400 micrograms twice a day; omeprazole, 20 mg once a day; or placebo, 1 tablet twice a day. Seven days after the end of the second source of CT, all patients underwent control esophagogastroduodenoscopy. Endoscopic findings were quantified on the basis of an arbitrary score: 0 = normal; 1 = less than 3 erosions; 2 = 3-15 erosions; 3 = more than 15 erosions or ulcer; 4 = giant ulcer (greatest dimension of more than 2 cm) or multiple ulcers with cumulative greatest dimension exceeding 2 cm.

RESULTS

Mean score increased significantly in the placebo and misoprostol groups, either after CMF (P < 0.001 and P < 0.05, respectively) or after 5-FU (P < 0.001 for both), whereas it did not in the omeprazole group. Gastric and duodenal ulcers were significantly less frequent in patients receiving omeprazole than in those receiving placebo (P < 0.05 after both CMF and 5-FU). No significant difference was observed between placebo and misoprostol. Omeprazole was significantly more effective than placebo and misoprostol in reducing the frequency and degree of the endoscopic worsening, either after CMF or after 5-FU (P < 0.05 for both CT regimens). Epigastric pain and/or heartburn were significantly less frequent in patients receiving omeprazole than in those receiving placebo (P < 0.01) or misoprostol (P < 0.001).

CONCLUSIONS

The strong and prolonged inhibition of gastric acid production induced by omeprazole seems to be effective in preventing chemotherapy-induced gastroduodenal mucosal injury. Further trials are necessary to verify whether such a prevention of endoscopically observed injury can translate into prevention of clinically significant injury.

Misoprostol and omeprazole in the prevention of chemotherapy-induced acute gastroduodenal mucosal injury. A randomized, placebo-controlled pilot study

BACKGROUND

Chemotherapy (CT) may induce acute mucosal injury to the stomach and duodenum, but its prevention has been scarcely investigated.

METHODS

One hundred and eighty-two cancer patients with normal stomach and duodenum or having fewer than 3 erosions, selected to be treated with cyclophosphamide, methotrexate, and 5-fluorouracil (CMF) (77 breast carcinoma patients) or 5-fluorouracil (5-FU) (105 colon carcinoma patients), were randomly assigned to prophylactic treatment with misoprostol, 400 micrograms twice a day; omeprazole, 20 mg once a day; or placebo, 1 tablet twice a day. Seven days after the end of the second source of CT, all patients underwent control esophagogastroduodenoscopy. Endoscopic findings were quantified on the basis of an arbitrary score: 0 = normal; 1 = less than 3 erosions; 2 = 3-15 erosions; 3 = more than 15 erosions or ulcer; 4 = giant ulcer (greatest dimension of more than 2 cm) or multiple ulcers with cumulative greatest dimension exceeding 2 cm.

RESULTS

Mean score increased significantly in the placebo and misoprostol groups, either after CMF (P < 0.001 and P < 0.05, respectively) or after 5-FU (P < 0.001 for both), whereas it did not in the omeprazole group. Gastric and duodenal ulcers were significantly less frequent in patients receiving omeprazole than in those receiving placebo (P < 0.05 after both CMF and 5-FU). No significant difference was observed between placebo and misoprostol. Omeprazole was significantly more effective than placebo and misoprostol in reducing the frequency and degree of the endoscopic worsening, either after CMF or after 5-FU (P < 0.05 for both CT regimens). Epigastric pain and/or heartburn were significantly less frequent in patients receiving omeprazole than in those receiving placebo (P < 0.01) or misoprostol (P < 0.001).

CONCLUSIONS

The strong and prolonged inhibition of gastric acid production induced by omeprazole seems to be effective in preventing chemotherapy-induced gastroduodenal mucosal injury. Further trials are necessary to verify whether such a prevention of endoscopically observed injury can translate into prevention of clinically significant injury.

Percutaneous endoscopic gastrostomy placement using the pull-through or push-through techniques: is the second pass of the gastroscope necessary?

BACKGROUND AND STUDY AIMS

The pull-through and push-through techniques widely used for placing a percutaneous endoscopic gastrostomy (PEG) require two passes of the gastroscope. The second pass is considered necessary to assess the correct positioning of the internal bumper. The aim of the present study was to verify whether the second pass is in fact necessary, or whether it could be omitted in most cases.

PATIENTS AND METHODS

Eighty patients undergoing pull-through or push-through PEG placement were included in this prospective study, and were randomly assigned to two groups. In the first group, two passes of the gastroscope were carried out, while in the second group the second pass was omitted and the position of the internal bumper was manually assessed by finger palpation of the abdominal wall after the feeding tube had been pulled out through the abdomen. If finger palpation was not considered satisfactory, a control gastroscopy was carried out. In the other cases, plain radiographs of the abdomen were carried out within six hours of the procedure.

RESULTS

Thirty-nine patients underwent two-pass PEG placement, and 41 received one-pass PEG placement. In one patient who had a one-pass procedure, the finger palpation was not considered satisfactory; however, a control gastroscopy showed that the internal bumper was correctly placed. In all other one-pass PEG patients, plain radiography of the abdomen showed that the internal bumper was at an adequate distance from the abdominal wall. No major or minor complications associated with the procedure were observed in either group. Bowel sounds reappeared within 24 hours of the procedure in all patients except for one in the two-pass group, in whom they reappeared after 36 hours. One wound infection, treated with systemic antibiotics, occurred in each group within 30 days of the PEG placement. The procedure time saved in the one-pass PEG group averaged 1.5 minutes.

CONCLUSIONS

One-pass PEG placement appears to be as safe as the classic pull-through and push-through PEG methods, and can be used routinely to make the procedure quicker and reduce the discomfort for the patient, as well as reducing the risks and costs associated with the second pass of the gastroscope. A control gastroscopy can be reserved for patients with morbid obesity, or for cases in which finger palpation is not considered satisfactory.